snr.control: Set Control Parameters for snr

In assist: A Suite of R Functions Implementing Spline Smoothing Techniques

Set Control Parameters for snr

Description

Control parameters supplied in the function call replace the defaults to be used in calling snr.

Usage

snr.control(rkpk.control = list(job = -1, tol = 0, init = 0, limnla = c(-10, 
    0), varht = NULL, theta = NULL, prec = 1e-06, maxit = 30), 
    nls.control = list(returnObject = TRUE, maxIter = 5), incDelta = 0.001, 
    prec.out = 0.001, maxit.out = 30, converg = "COEF", method = "GN", 
    backfit = 5) 

Arguments

rkpk.control	a optional list of control parameters for dsidr or dmudr to estimate the unknown functions. Default is "list(job = -1, tol = 0, init = 0, limnla = c(-10, 0), varht = NULL, theta = NULL, prec = 1e-06, maxit = 30)".
nls.control	a list of control parameters for the nonlinear regression step, the same as gnlsControl. Default is "list(returnObject = TRUE, maxIter = 5).
incDelta	the incremental value to be used to calculate derivatives for the unknown functions. Default is 0.001
prec.out	tolerance for convergence criterion. Default is 0.0001.
maxit.out	maximum number of iterations for the algorithm. Default is 30.
converg	an optional character, with possible values COEF and PRSS, specifying the convergence criterion to be used. COEF uses the change of estimate of parameters and functions to assess convergence, and PRSS uses penalized residual sums of squares. Default is COEF.
method	an optional string of value either GN for Gauss-Newton or NR for Newton-Raphson iteration methods to estimate the unknown functions. Default is GN.
backfit	an integer to set the number of backfitting iterations inside the loop. Default is 5

.

Value

returned is a list includes all re-seted control parameters.

Author(s)

Chunlei Ke [e-mail] and Yuedong Wang [e-mail].

See Also

snr, dsidr, dmudr

Examples

## use Newton-Raphson iteration and only a single backfitting
## Not run: 
snr.control(method="NR", backfit=1)

## End(Not run)

assist documentation built on Aug. 22, 2023, 9:12 a.m.